Design of artificial neural networks for darcy-forchheimer flow of boger hybrid nanofluid with CattaneoChristov theory: Thermofluidic applications

Abstract

This study investigates the effects of Marangoni convection on the Darcy-Forchheimer flow of MHD Boger fluid across a sheet using CattaneoChristov heat and mass flux model using artificial neural networks and the Levenberg–Marquardt scheme. The proposed artificial neural network model for Darcy-Forchheimer flow of Boger hybrid nanofluid using CattaneoChristov theory has important thermofluidic applications in engineering and industrial processes that require accurate heat and mass transport management. Its ability to capture non-Fourier heat conduction and resist porous media makes it ideal for optimizing cooling in high-temperature electronics, improving heat exchangers in geothermal and solar-thermal systems, improving thermal management in energy storage devices, and designing efficient thermal barrier coatings. The use of artificial neural networks allows for efficient prediction and optimization of flow and heat transfer characteristics, saving computational time and enabling real-time control in engineering applications. Similarity variables are used to transform nonlinear partial differential equations into nonlinear ordinary differential equations. Numerical results are simulated using the Bvp4c. The approximate solutions for each case are then analyzed using Levenberg Marquardt artificial neural networks for testing, training, and validation. The Levenberg Marquardt artificial neural networks are validated by regression studies, histogram analysis and mean square error. The results are shown to show the outcome of different physical conditions on the associated distributions.